Sacred groves hold distinct bird assemblages within an Afrotropical savanna

Global Ecology and Conservation 18 (2019) e00656

Contents lists available at ScienceDirect

Global Ecology and Conservation journal homepage: http://www.elsevier.com/locate/gecco

Original Research Article

Sacred groves hold distinct bird assemblages within an Afrotropical savanna Katharina Kühnert a, *, Ingo Grass b, Matthias Waltert a € t Go €ttingen, Workgroup on Endangered Species, J.F. Blumenbach Institute of Zoology and Anthropology, Georg-August-Universita €ttingen, Germany Bürgerstr. 50, 37073, Go b €t Go €ttingen, Grisebachstr. 6, 37077, Go €ttingen, Germany Department of Crop Science- Agroecology, Georg-August-Universita a

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 December 2018 Received in revised form 6 May 2019 Accepted 7 May 2019

Riparian forests, an integral part of savanna ecosystems, are threatened across West Africa by agricultural expansion. However, some patches of original riparian vegetation are protected by traditional beliefs as ‘Sacred Groves’. We assessed the role of Sacred Groves in maintaining landscape-scale bird assemblages by conducting 144 1-h point counts, distributed over 24 plots in eastern Guinea-Bissau. The plots were situated in three riparian habitat types with different levels of human modification (Sacred Grove, Young Secondary Forest, Annual Cultures) and the adjacent Wooded Savanna. We accumulated 4572 records of 174 species and compared total species richness, composition, and functional traits among the four habitat types. At the plot level, species richness was higher in Wooded Savannas and Annual Cultures compared to Secondary Forests and Sacred Groves. Bird communities in Wooded Savannas were similar to those in Annual Cultures and differed the most from those of Sacred Groves. Bird community composition in Young Secondary Forests was similar to that in Annual Cultures but showed a shift towards the community composition found in Sacred Groves. Certain traits were strongly specific to habitat type. For example, Sacred Groves were characterized by a high number of forest specialists and insectivorous birds. Our results suggest that the rapid successional dynamics in riparian habitats enable disturbance tolerant forest species to recolonize fallow areas after a relatively short period of time. However, Sacred Groves hold a distinct avifauna and their conservation may therefore be crucial for forest specialist species and the re-establishment of bird assemblages in fallow riparian areas. Our findings also stress the importance of respecting and strengthening traditional forms of nature protection. © 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Fourth corner analysis Functional diversity Guinea savanna biome Guinea-Congo biome Land use Sacred groves

1. Introduction Many societies traditionally possess sites set aside for spiritual rituals and their deities that guard and provide for their communities (Bhagwat and Rutte, 2006). Examples of sacred sites include monastery gardens in Europe (Helms, 2002), sacred caves in Kazakhstan (Lymer, 2004), Indian Sacred Groves (Ormsby and Bhagwat, 2010) and Sacred Groves in the coastal savannas of Ghana (Campbell, 2005). Numerous case studies have demonstrated the value of Sacred Groves for biodiversity conservation (Decher, 1997; Ormsby and Bhagwat, 2010; Aerts et al., 2006).

* Corresponding author. E-mail addresses: [email protected] (K. Kühnert), [email protected] (I. Grass), [email protected] (M. Waltert). https://doi.org/10.1016/j.gecco.2019.e00656 2351-9894/© 2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

2

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

The relevance of Sacred Groves as a conservation tool is augmented in places where governmental protection of natural resources is weak or non-existent. This is the case in most of West Africa, a highly biodiverse region with relatively fragile political stability and poor law enforcement (Maconachie et al., 2015). The existence of Sacred Groves in the region has been described in a number of studies (e.g. Lebbie and Freudenberger, 1996; Oteng-Yeboah, 1996; Campbell, 2005; Ceperley et al., 2010). The conservation of these traditionally protected sites becomes more important as forests vanish (Barre et al., 2009). Deforestation took place across the whole of West Africa during the last decades, caused by the demand of a growing human population and associated land use changes (Brink and Eva, 2009; Goetze et al., 2006; Jalloh et al., 2012). Decreases in forest cover have been particularly rapid in the transition zone between the Guinean forest and the Sudanian savanna, where about 20% of forest cover has been lost between 1975 and 2000 (Brink and Eva, 2009). In this transition zone, the landscape consists of multiple habitat types: grasslands cover large areas, interspersed with woody vegetation ranging from isolated bushes to small islands of dry forest. Evergreen riparian forests constitute additional conspicuous features in this landscape. These riparian forests disappear faster than other forest types, mainly being converted for agriculture (Goetze et al., 2006). The effects of these changes on biodiversity remain largely undocumented, although general and severe declines across taxa have been described in the region (Mallon et al., 2015). For example, strong population declines in both migratory and non-migratory birds have been reported (Sanderson et al., 2006; Thiollay 2006a; Vickery et al., 2014). These declines are caused by a combination of factors which include climate change (Møller et al., 2008), increased hunting pressure (e.g. Trail, 2007; Whytock and Morgan, 2010; Whytock et al., 2016), and deliberate poisoning (Ogada et al., 2012). Nevertheless, land use changes and forest disturbance can be considered the main causes for the decline of bird populations in the region (Arcilla et al., 2015; Waltert et al., 2005). The effects of forest disturbance and modification on bird communities are well known. Previous research shows that feeding guild is a good predictor of disturbance sensitivity in tropical bird communities (Gray et al., 2007). For instance, insectivores are very sensitive to tropical forest disturbance, mainly due to limitations in their dispersal ability (S¸ekercioḡlu et al.,2002). For large frugivorous species, body size can serve as an indicator for vulnerability to the indirect effects of forest disturbance, such as increased hunting pressure or selective logging (Markl et al., 2012). In contrast, granivorous species usually thrive in areas cleared for smallholder agriculture (Gray et al., 2007; Wiens and Johnston, 2012). Although these drivers are well understood for tropical forest landscapes, comparatively little is known from savanna landscapes in West Africa, where dry forests and riparian forests are patchily distributed. In eastern Guinea-Bissau, the last remaining patches of primary vegetation in riparian areas are conserved as Sacred ^te d’Ivoire (Goetze et al., 2006) and Benin Groves. This mirrors the situation in other countries in the region, such as Co (Ceperley et al., 2010). Yet pressures on Sacred Groves are high due to an increased demand for arable land, driven by population growth and the spread of Cashew as a cash crop (Temudo and Abrantes, 2014). This pressure, in combination with a weakening of traditions and beliefs can lead to the abandonment and subsequent cutting of Sacred Groves for agriculture. € derstro €m et al., 2003;), effects of riparian forest loss on Although small scale agriculture may benefit some bird species (So overall bird assemblage remain largely undocumented. Moreover, little is known on the role of Sacred Groves in reducing the impacts of riparian forest loss in West African savannas (but see Ceperley et al., 2010). To address these knowledge gaps, we  sector of eastern here analysed the role played by Sacred Groves in maintaining landscape-scale bird assemblages in the Boe Guinea-Bissau. We studied the bird communities of Sacred Groves (representing near-primary riparian forests) and the adjacent Wooded Savannas. To better understand the effects of riparian forest modification, we also studied the bird communities of modified habitats located in riparian areas, i.e. Annual Cultures and Young Secondary Forests. These modified habitats represent different successional stages of formerly forested riparian area. Given that savannas cover a comparably larger area in the study region (Wit and Reintjes, 1989), and across Africa (White, 1983), we expected that Wooded Savannas support a higher number of bird species compared with the riparian habitat types (Sacred Grove, Young Secondary Forest, Annual Culture) (Rosenzweig, 1995). We expect that bird community composition clearly differs between the structurally more open Wooded Savannas and closed Sacred Groves, as they represent very distinct habitat types. We predict that the bird assemblages of modified riparian habitat types (Annual Culture and Young Secondary Forest) are intermediate to Sacred Grove and Wooded Savanna. Finally, we anticipated to find trait-related effects in the composition of the bird community, associated with different stages of succession. In particular, we expected a loss of forest specialist species and insectivores in the Annual Cultures and Young Secondary Forests (Waltert et al., 2005; Arcilla et al., 2015). In contrast, we expected granivorous species to be positively associated with Annual Cultures (Wiens and Johnston, 2012). 2. Methods 2.1. Study area and plots  region in eastern Guinea-Bissau (Fig. 1) is situated between 11
300 and 12
05 0 northern latitude and between The Boe 13
45 0 and 14
30 0 western longitude, having a tropical climate and an annual precipitation between 1600 and 2100 mm. The  region is covered by a hardpan of laterite rock, allowing only grasses to establish (Wit and Reintjes, major part of the Boe 1989). In depressions, where soil accumulates, islands of dry forest are common. The numerous streams of the region were originally framed by evergreen riparian forests, which are now almost completely cut for shifting cultivation.

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

3

Fig. 1. Study area. To the left, a map indicating the location of the study area in eastern Guinea-Bissau. To the right, a map of the distribution of study plots across the study area, with the original shape and extent of Sacred Groves in white, Annual Cultures are marked with a flag, Young Secondary Forests represented by a star, and Wooded Savannas by a grey circle. 1.5 column fitting image, printed in color only for online version.

Study plots were chosen non-randomly due to the geographical characteristics of the study region, being distributed over li. The average distance between any two plots was 5 km, with a an approximate area of 100 km2 around the village of Be minimum distance of 470 m. Sacred Groves represent the last remaining patches of unlogged riparian forest, often situated at the origins of streams or rivers, comparable to the ones described for Benin by Ceperley et al. (2010). The average size of Sacred Groves visited was 2.8 ha, except from “Bundu Njuuri”, which was exceptionally large (29 ha). This larger forest patch is still used for rituals as they were commonly carried out in sacred sites in the past (pers. comm, Bucari Cassama, 2015). For this reason, “Bundu , Njuuri” is possibly better protected by traditional beliefs than the other Sacred Groves in our study (pers. comm, Mussa Sane 2015).  is similar to the slash-and burn practices found across West Africa. In the Boe , The subsistence agriculture in the Boe Annual Cultures are often situated in stream or river valleys and forested buffer zones (ranging from single trees to a few , 2015). meters of forest stand) are usually kept on the valley borders to act as fire- and windbreaks (pers. comm., Mussa Sane However, due to the low soil quality, the same field can only be cultivated for a maximum of two consecutive years, and is , 2015; Temudo and Santos, 2017). The plots classified as Annual then left fallow for 4e7 years (pers. comm.,Mussa Sane Cultures corresponded to fields which had been cultivated for rice, millet, and sorghum in 2014 and had been left fallow for about 4 months before the start of our study. Only one plot had been cultivated again in 2015 (Annual Culture “Quinhique”, Table A1). All Annual Culture plots were situated in river valleys, except from one, which was situated on a valley border li”, Table A1). This location was chosen due to the limited number of suitable Annual Culture plots within (Annual Culture “Be an accessible distance, because most riverine areas close to the village are cultivated for with cashew. Young Secondary Forest plots chosen for our study were found in areas that had been previously cultivated and had been left fallow for 4e6 years, allowing young trees to establish with an approximate height of 3e5 m (Fig. 2). Information on the age of Young Secondary Forest plots was retrieved from the local guides and, if necessary, confirmed by the owner of the plot  and Balu Se ra, 2015). (pers. comm., Mussa Sane Finally, locations of Wooded Savanna plots were chosen based on their logistic suitability, as all plots had to be accessible by bicycle.

4

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Fig. 2. Successional chrono-sequences among studied habitats. The arrows (1e4) indicate the successional habitat chrono-sequence as follows: 1: Riparian forest (i.e. Sacred Grove, n ¼ 6, top left) is cut and cultivated in Annual Culture (n ¼ 6, bottom left). 2: Annual Culture is left fallow for 4e6 years and develops into Young Secondary Forests (n ¼ 6, top right). 3: In theory, Young Secondary Forests could develop into mature riparian forests again, a process that is unlikely given high population pressure. 4: Instead, Young Secondary Forests most often gets cut again after 4e6 years of fallow and become Annual Cultures. Adjacent Wooded Savanna is displayed at the bottom-right (n ¼ 6). All pictures were taken in March 2015. 1.5 column fitting image, printed in color only for online version.

2.2. Bird counts Fixed-radius point counts were carried out by KK during the morning between 06h30 and 12h00 from March to July 2015. During 1 h, observations of individual birds were recorded only if the possibility of double-counting could be excluded. To account for time and temperature effects on counting success, counts were carried out twice during early, mid and late morning, totaling six visits per plot. All counts were done under dry conditions, with 2e3 weeks between each visit. Only detections within an estimated distance of 50 m were considered for analysis (Sutherland, 2006). Birds were detected either visually or acoustically, after waiting for 3 min quietly in the center of the plot. 2.3. Vegetation and GIS survey Vegetation structure estimates were done twice; once at the beginning of the study period and once at the end of it, to account for the level of structural changes during the five-month study period (Table A.2). Percentage of vegetation cover was visually estimated within a 50 m radius around the center of each plot on four different height bands. Height bands were placed at 1 m, 2 m, 4 m, 8 m, and 16 m, to account for variation in height structure among the investigated habitat types. The 50 m radius was quartered, resulting in four estimates per plot and height band. Canopy cover estimates were retrieved from the center of each quarter by using a sighting tube. All estimates were taken by the same observer and in accordance with Bibby et al. (2000). To retrieve an index of vegetation height heterogeneity, Shannon diversity was calculated by using estimates of vegetation cover for each height band and per plot. Estimates were averaged over the four plot quarters prior to calculation. For final analyses, estimates from the beginning and end of the study period were averaged and are referred to as “vegetation height heterogeneity” hereafter. To retrieve a measure of landscape diversity for each plot, we mapped habitat composition within a 200 m radius around each plot in ArcGIS (Morelli et al., 2013). By using satellite imagery (Google Earth, 2015) we distinguished five habitat categories corresponding to the habitat types investigated: Old growth forest (corresponding to Sacred Groves), fallow (corresponding to Young Secondary Forests), savanna (corresponding to Wooded Savanna, but also including open savanna areas), cultivation (Annual Cultures), and village (no correspondence). The latter category was included, because the 200 m radius of few plots stretched into village area. Almost all plots included shares of all five habitat categories, so we were unable to use category richness as an indicator, but instead calculated category diversity (Morelli et al., 2013). We calculated the Shannon-Wiener index based on percentage shares of the prementioned classification types within a 200 m radius around each plot (Morelli et al., 2013).

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

5

2.4. Species traits Species trait data were retrieved from Borrow and Demey (2008, 2011), Fry and Fry (2010), and the IUCN (2015). We focused on six traits: feeding guild, migratory behavior, body mass, habitat guild, biome association and global population trend (Table A.3). Birds were divided into feeding guilds based on their primary feeding behavior: carnivorous, insectivorous, frugivorous, nectivorous, granivorous, and omnivorous (Borrow and Demey, 2008, 2011). Moreover, species were classified into resident, breeding visitor, or nonbreeding visitor since migratory birds might respond differently to habitat modification than resident species (e.g. Sanderson et al., 2006). Our dataset contains four species of Palearctic migrants: European Beeeater Merops apiaster, Eurasian Golden Oriole Oriolus oriolus, Sand Martin Riparia riparia, and Common Swift Apus apus, which were not treated separately, but classified as nonbreeding visitors. To gain further information on the conservation value of the studied habitat types, information on biome-association was added for each species, following Fishpool and Evans (2001). Bird species were either restricted to the Guinean-Congo Forest biome, the Sudan-Guinean Savanna biome or non-restricted. Finally, the global population trend of each species was added, aiming to test whether increasing, decreasing, stable or unknown global population trends are associated to any of the studied habitat types (IUCN, 2015). 2.5. Data analyses The number of species detected after six repeated visits was counted for each plot and is referred to as “observed species richness” hereafter. In total, 224 bird species were detected, out of which only 174 species were used for analysis, because all observations beyond 50 m distance were omitted. Normally, some species are missed during a field study (Nichols and Conroy, 1996), so to gain a more accurate picture of the actual number of species, we calculated an ‘estimated’ species richness, using the R package vegan (Oksanen et al., 2015), with the incidence-based and bias-corrected Chao 1 estimator (Chao, 1987). This method infers the number of missed species from the frequency of species in a collection of plots (Palmer, 1990; Colwell and Coddington, 1994). All further analyses were undertaken at plot level, i.e. we used the accumulated number of birds after six repeat visits per plot (n ¼ 24 plots). To test for differences in species richness among habitat types, a general linear model (GLM) with Poisson distribution for count data was used, with vegetation height heterogeneity, canopy cover, and landscape diversity as covariates. To assess differences in species composition among habitat types, a permutational multivariate analysis of variance (PERMANOVA) was carried out, using a dissimilarity matrix, based on Bray-Curtis distances. This distance metric is insensitive to changes in absolute abundances, which is of importance for our data, since absolute abundances strongly differed among habitat types (Anderson, 2001). Based on the dissimilarity matrix used for PERMANOVA, species community composition was more closely examined, using non-metric multidimensional scaling (NMDS). NMDS uses rank orders of species for ordination, making data less sensitive to transformations which are needed to obtain a meaningful ordination. Our data were standardized by square root and Wisconsin transformation. NMDS predicts values for each species per plot in a multidimensional space (one dimension per species). Then, it calculates the best two-dimensional solution for ordination, indicated by the lowest stress value (here: 0.138). The stress value quantifies the discrepancy between predicted values and values assigned to each data point by fitting it to a two-dimensional solution for displaying purposes. Species composition analysis was carried out using the R package vegan (Oksanen et al., 2015). To investigate the effects of habitat and environmental variables on specific species traits, we used fourth corner analysis, model-type 1 with unadjusted p-values (Dray and Legendre, 2008). We included the following environmental variables: habitat type, vegetation height heterogeneity, canopy cover, and landscape diversity. Species-associated trait categories were: habitat, feeding, and migratory guild, biome affiliations, and the IUCN Red List of Threatened Species’ population trend (IUCN, 2015). Since the fourth corner analysis is an abundance-based method, it could potentially yield stronger effects for plots with higher abundances. To preclude such a bias, we divided species-specific abundance by total abundance per plot prior to analysis. Hence, the fourth corner analysis was carried out using dominances per plot, rather than absolute abundances. Fourth corner analysis was done using the R package “ade4” (Dray and Dufour, 2007). All statistical analyses were done in R 3.2.2 (R Core Team, 2015). 3. Results 3.1. Bird species richness and diversity In total, 174 bird species, accounting for 4572 individuals were analysed (Table A.3). Based on this, we estimated a joint total of 209 (±17 SE) species (Chao1 estimator) across all studied habitat types and drew habitat-specific species accumulation curves (Fig. A1). According to species richness estimates, roughly, 87% of species were detected in Sacred Groves, 75% in Young Secondary Forests, 85% in Annual Cultures, and 78% in Wooded Savannas. The highest number of species was observed in Annual Cultures, followed by Wooded Savannas, Young Secondary Forests, and Sacred Groves (Fig. 3). All further results refer to the plot level (n ¼ 24 plots) and have been obtained using the accumulated number of observed bird species after six visits for each plot. The GLM analysis revealed, that observed species richness was determined by habitat

6

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Fig. 3. Observed (black bars) and estimated (grey bars) bird species richness in each studied habitat type. Estimated species richness is based on Chao estimator (black lines indicating the standard error). Different letters (a, b, c, d) indicate whether habitats significantly differ in species richness, using a linear model with vegetation measures and landscape diversity as covariates (LM, P < 0.05). SAGR¼Sacred Grove, YSEC¼Young Secondary Forest, ANC ¼ Annual Culture, SAV¼Wooded Savanna 1 column fitting image.

type. Observed species richness was highest in Annual Cultures (Z ¼ 5.164, P < 0.001), followed by Wooded Savannas (Z ¼ 3.287, P ¼ 0.001), and Young Secondary Forests, where habitat type had only a marginal influence on species richness (Z ¼ 1.686, P ¼ 0.091). Species richness in Sacred Groves however, was significantly determined by habitat type (Z ¼ 11.549, P < 0.001) (Fig. 3). Observed species richness in Annual Culture plots was significantly higher compared with Young Secondary Forest plots (Tukey's HSD, P ¼ 0.001) and plots in Sacred Groves (Tukey's HSD, P < 0.001), but did not significantly differ from the species number observed in Wooded Savanna plots (Tukey's HSD, P ¼ 0.84) (Fig. 4). Observed species richness in Wooded Savannas was, in turn, only different from Sacred Groves (Tukey's HSD, P ¼ 0.005) (Fig. 4). The forested habitat types, i.e. Sacred Groves and Young Secondary Forests, did not show a difference in observed species richness on plot level (Tukey's HSD, P ¼ 0.32) (Fig. 4). Running a GLM with the same covariates as above, estimated bird species richness (Chao1 estimator) showed a very similar pattern. But for estimated species richness, Wooded Savanna plots held significantly more species than Young Secondary Forest plots (Tukey's HSD, P < 0.001). Shannon bird diversity did not differ between habitat types on the plot level (Fig. 4). 3.2. Effects of vegetation variables and landscape diversity No difference in vegetation height heterogeneity (one-way ANOVA, P ¼ 0.7) nor canopy cover (one-way ANOVA, P ¼ 0.3) was detected at plot level. Nonetheless, vegetation height heterogeneity had a marginally negative effect on observed species richness (P ¼ 0.077), but a significantly negative effect on estimated species richness (P ¼ 0.003). Both, observed and estimated species richness decreased with increasing canopy cover (Poisson GLM, P ¼ 0.033; and GLM, P < 0.001 respectively). Landscape diversity differed among habitat types (P < 0.001), being particularly lower for Wooded Savanna plots compared to all other habitat types (Tukey HSD, P < 0.05 in all cases). However, landscape diversity had no effect on observed and estimated species richness. 3.3. Community composition We found a significant effect of habitat type on bird community composition, which was revealed by PERMANOVA test (R2 ¼ 0.34, P < 0.050). All other covariates (i.e. vegetation height heterogeneity, canopy cover and landscape diversity) did not significantly affect community composition. NMDS ordination visualized differences in bird community composition among habitat types at the plot level (NMDS, stress ¼ 0.13, Fig. 5) and showed a clear divergence in bird assemblages of open habitat types (i.e. Wooded Savanna and Annual Culture) compared with closed habitat types (i.e. Sacred Grove and Young Secondary Forest). We observed an overlap in community composition between Annual Cultures and Wooded Savannas. However, this

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

7

Fig. 4. Plot level biodiversity estimates. Observed and estimated species richness and Shannon diversity for Sacred Groves (SAGR, n ¼ 6), Young Secondary Forests (YSEC, n ¼ 6), Annual Cultures (ANC, n ¼ 6), and Wooded Savannas (SAV, n ¼ 6). Different letters (a, b, c) indicate, whether habitats significantly differ in the respective measure (P < 0.05). Open dots mark outliers. 1.5-column fitting image.

overlap was caused by a single Annual Culture plot (Fig. 5). Plots of Young Secondary Forests and Sacred Groves were adjoining, but non-overlapping. 3.4. Trait-specific responses We looked for trait-related effects which drive differences in community composition between habitat types. Fourthcorner analysis revealed significant relationships among certain species traits, environmental variables and habitat type (Table 1). For example, the bird community of Sacred Groves was positively associated with species belonging to habitat guilds characteristic to primary and secondary forest but negatively associated with species of rank herbage-/moist habitats, and (wooded) savannas (Table 1). Carnivorous, insectivorous, and nectivorous species were positively associated, while granivorous species were negatively correlated with Sacred Groves. Also, a positive association was found between Sacred Groves and Guinean-Congo forest biome-restricted species (Table 1).The bird community of Young Secondary Forests showed a comparatively higher number of frugivorous birds and was represented by species commonly described to favor secondary forests (Table 1). Birds described to live in savannas and wooded savannas were underrepresented in Young Secondary Forests but positively associated with Annual Culture - and Wooded savanna plots (Table 1). In contrast, Guinean-Congo forest biome restricted species were negatively associated with the open habitat types Annual Culture and Wooded Savanna. Birds typically found in secondary forests and forests were also still supported by the open habitat types of our study, i.e. Annual Culture and Wooded Savanna (Table 1). Annual Cultures were negatively associated with insectivorous bird species and positively with granivorous species (Table 1). Migratory birds, categorized as breeding visitors or non-breeding visitors, were positively associated with Wooded Savanna plots. Resident species were positively associated with landscape diversity (Table 1). None of the investigated habitat types was associated with a particular IUCN population trend, nor any of the environmental indicators (Table 1). 4. Discussion Sacred Groves help to maintain overall bird assemblages in the Sudan-Guinean savanna transition zone. The community composition of Sacred Groves was distinct from the bird assemblage of Wooded Savannas, but this distinction faded in modified riparian habitat types, with bird assemblages being intermediate to Sacred Groves and Wooded Savannas. We also demonstrated guild specific responses to habitat modification, with e.g. a loss of insectivorous species in Annual Cultures and Young Secondary Forests. In contrast, Sacred Groves of our study significantly support carnivorous, nectivorous and insectivorous species, despite a comparably low species richness. The low species richness of Sacred Groves observed in our study contrasts other studies which find higher species richness in Sacred Groves compared with the surrounding landscape (e.g. Brandt et al., 2013). The higher species richness of Wooded

8

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Fig. 5. Non-metric-multidimensional scaling ordination of the 24 plots sampled across four habitat types (Sacred Groves: dark green, Young Secondary Forest: light green, Annual Cultures: red, Wooded Savanna: yellow), based on Bray-Curtis distances of bird abundances. Plots belonging to the same habitat are connected by lines or fall within these lines. The chrono-sequence of habitat succession is additionally illustrated by arrows 1 to 4. 1: Riparian forest (i.e. Sacred Grove) is cut and cultivated (becoming Annual Culture). 2: Annual Culture is left fallow and develops into Young Secondary Forests. 3: Young Secondary Forests potentially develops into mature riparian forests 4: Instead, Young Secondary Forests get cut again after four to six years of fallow and become Annual Cultures again. 1.5 column fitting image, color printed only in online version.

Table 1 Fourth corner analysis of the communities of each investigated habitat in relation to environmental variables and species traits. Positive (þ) and negative () associations for a significance threshold of P < 0.05 are calculated based on Pearson correlation. Traits Trait class

Habitat types Trait Category

Habitat guild Forest Secondary Forest Wooded Savanna Savanna Rank Herbage/Moist Habitat Feeding guild Carnivorous Insectivorous Omnivorous Granivorous Frugivorous Nectarivorous Body size Migratory Resident guild Breeding visitor Nonbreeding visitor Biome Guinean-Congo Association Forest Sudan-Guinean Savanna None

Environmental Measures

Sacred Groves

Young Secondary Forest

Annual Cultures

Wooded Savannas

þ þ e e e

þ e e

e e þ þ þ

e e þ þ

þ þ

Vegetation height heterogeneity

Canopy cover

Landscape diversity þ þ

þ

þ

e

e

þ þ

þ þ e þ

e

e

þ þ

þ þ e

þ

þ

þ

Savannas might be attributed to several factors acting at different spatial scales. Our study was set up in a context where savannas represent the dominating habitat type (Wit and Reintjes, 1989). Savannas also represent the dominating habitat type across Africa (White, 1983), so according to the species-area relationship, we could expect a higher species richness in savannas, compared to other habitat types (Rosenzweig, 1995). On a landscape level, the isolated patch character of Sacred Groves (Bender et al., 1998), combined with the limited width of river valleys might restrict avian species richness therein (e.g. Kinley and Newhouse, 1997; Shirley and Smith, 2005). At local level, Decher (1997) found a similar pattern for small mammals in Sacred Groves of Ghanaian savannas: although species richness of bats was lower in Sacred Groves, it comprised forest

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

9

specialist species uniquely observed there. Hence, despite being comparably species-poor, community composition of Sacred Groves was unique. Also in our study, the bird communities between the near-natural habitat types Sacred Grove and Annual Culture were distinct. As expected, bird community composition of modified riparian habitat types (i.e. Annual Culture and Young Secondary Forest) was intermediate to the near-natural habitat types Sacred Grove and Wooded Savanna. Although Young Secondary Forests had been left fallow during only four to six years, the bird community had already shifted from one typical of Annual Cultures towards one that closely resembled Sacred Groves. Such a rapid convergence of the avian community exceeds that reported in other studies (Raman et al., 1998; Naidoo, 2004), and may have been facilitated by the exceptional conditions in riparian zones which accelerate vegetation growth (Gregory et al., 1991). More likely, though, this rapid convergence documents an already highly impoverished avifauna of Sacred Groves, being little different from Young Sec region must have been gradually deforested since humans settled, with ondary Forests. The original riparian forests in the Boe little more left today than small patchily distributed Sacred Groves. Likely, extinction events took already place and might still be ongoing (e.g. Brooks et al., 1999; Brooks et al., 2002). For example, there is anecdotal evidence on the earlier occurrence of the Great Blue Turaco Corythaeola cristata (pers. comm., Amadal Camara, 2015). However, the high species richness in Annual Cultures proves that cultivated riparian areas can support a high number of bird species. A similar pattern is known from Burkina Faso, where avian species richness was highest in cultivated areas and € derstro €m et al., 2003). Conversely, Thiollay (2006b) discusses a species decreased as the fallow became more “forest-like” (So decline in cultivated areas compared to surrounding Savannas in Burkina Faso. These different results may be because Thiollay (2006b) focused on large-bodied birds, but we also included small, granivorous species. Our results showed that granivorous species were particularly common in Annual Cultures. Wiens and Johnston (2012) described that granivorous species are particularly successful in agricultural areas, where seeds are seasonally highly abundant, and we observed several species feeding on grains left from cultivation. Moreover, the open structure of Annual Cultures facilitates the colonization by species of savanna habitats which are to a large proportion small, seed-eating species. We conclude, that traditional culti, can provide a viable habitat for a great number of bird species (e.g. vation practices, as they are still carried out in the Boe Selmi et al., 2002; Mandal and Shankar Raman, 2016), although under the exclusion of forest specialists and insectivores. Alarmingly, these traditional cultivation practices are under threat from the expansion of Cashew as a cash crop (Temudo and Abrantes, 2014). Annual Cultures are commonly transformed into Cashew plantations which interrupts the fallow cycle (Temudo and Santos, 2017) and has a landscape level impact on biodiversity (Monteiro et al., 2017). The role of Sacred Groves as an important component for the protection of biodiversity at landscape level has received little attention so far (see Bhagwat et al., 2005; Sreevidya et al., 2016). In contrast, the role of landscape structure as a driver of bird diversity is well studied (Tscharntke et al., 2008). However, we were unable to detect such an effect in our study. This might be attributed to the small landscape radius investigated (200 m), and an inadequate data resolution. Furthermore, data on landscape diversity were skewed due to the comparable small-scale mosaic of modified riparian areas contrasting with vast savannas, far from rivers, cultivations, and settlements. As a result, all riparian habitat types showed a consistently high landscape diversity in contrast to Wooded Savannas. Finally, landscape effects might simply be negligible in the semi-natural mosaic landscape of our study area. Here, and the effects acting at the plot level might be more important. At the plot level, we observed a negative relationship between vegetation structure, canopy cover, and bird species richness, i.e. we found fewer species in structurally heterogeneous plots with higher canopy cover, compared to more simply structured open plots. These results are contrasting those from the majority of studies (MacArthur and MacArthur, 1961; Willson, 1974; Tews et al., 2004), yet are in accordance with the overall picture of a generally species rich Wooded Savanna and a poorer avifauna in the small, €derstro € m et al. (2003) from Burkina Faso. isolated Sacred Groves therein. This pattern is also confirmed by the study of So We finally want to point out, that highly forest dependent species, such as ant followers, are missing in the bird assemblages of our study. By nature, fewer forest specialist species will persist in the forest-savanna transition zone than in the zonal Guinean-Congo forest biome and they will be rarer. Biome transition zones, however, have been attributed a crucial role for processes of speciation and adaptation (Smith et al., 1997), indicating an important role of forest habitats in savanna regions. For example, Little Greenbuls Andropadus virens in the Sudan-Guinean forest-savanna transition zone, show songand morphological divergence compared to the zonal populations (Slabbekoorn and Smith, 2002). This phenotypic divergence has already decreased due to high deforestation across West Africa, reducing the species' adaptive diversity (Freedman et al., 2010). We conclude that the persistence of Sacred Groves in eastern Guinea Bissau, despite their comparably low bird species richness, is vital for long-term conservation efforts. Apart from forest specialist bird species, primates of the region highly dependent on riparian forests (Binczik et al., 2017) and thereby on Sacred Groves. Moreover, findings from other regions suggest that Sacred Groves might serve as source habitat for species to recolonize or survive in modified habitat types (Gao et al., 2013; Ray et al., 2015). Finally, Sreevidya et al. (2016) pointed out that the benefits provided by Sacred Groves go far beyond species protection. For example, Sacred Groves might be of high importance for watershed protection and water availability (Ray et al., 2015; Vinay et al., 2017), which is a burning issue for the human population in our study region. Our study offers the first comparative results on bird communities of modified riparian habitats and surrounding savannas in West Africa. Moreover, we add an insight into the conservation value of Sacred Groves, which are under threat in eastern Guinea-Bissau. Formal “conservation” however, holds some risks if initiated by outside institutions. A well-documented example of how outside interventions can weaken traditional forms of protection is the degradation of Sacred Groves in Cantanhez National Park (Guinea-Bissau), caused by shifts in power and a concurrent loss of perceived land security among

10

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

local communities (Temudo, 2012). Official policies may hence weaken traditional forms of protection of Sacred Groves (Brandt et al., 2015), if put in place without consideration of local power relations, beliefs and traditions. An effort to put the  region under formal protection is currently underway (Chimbo Foundation, 2019). We support this Sacred Groves of the Boe initiative, with a strong focus on local communities. The understanding of local beliefs and power relations is crucial for this kind of endeavor, and the implementing organization has more than a decade of experience in actively engaging locals into  (Chimbo Foundation, pers. comm). Therefore, we trust that any efforts will respect and conservation efforts in the Boe  region. consider local power relations and lead to a long-term protection of the Sacred Groves in the Boe Declaration of interests None. Acknowledgments This work was supported by the German Society for Tropical Ornithology (GTO), AKB Foundation. We acknowledge €ttingen University. We also thank support by the German Research Foundation and the Open Access Publication Funds of Go the Chimbo Foundation for logistical and technical support in the field. We are particularly grateful to Annemarie Goedra for his sharp makers, Piet Wit, Brecht Coppens, and Miguel Lecoq for support during this study. We want to thank Balu Se  for his priceless help and an open mind. We also want to thank Paula Roig eyes and endurance in the field and Mussa Sane Boixeda for her valuable input and efforts for the improvement of this manuscript. All the research conducted under this study complies with the law of the country where it was performed. We acknowledge support by the German Research €ttingen University. Foundation and the Open Access Publication Funds of the Go APPENDIX

Table A.1 Information on studied plots. Name, habitat type, location (WGS 1984; UTM Zone 28N), and elevation of each point count location visited during the study in  region (Guinea-Bissau). the Boe Plot name

Habitat type

X coordinate (UTM)

Y coordinate (UTM)

Elevation (m)

Cael Quebube2 li Be Quebube1 Queco Qunhique Bantandjan Bandugada Bundujuri Horegudum Horejalede Bundugada Horeferanora Belisinho Cael Pataque Quebube Queco Quebube1 Bundjuri Quebube2 Unduferanora Bantandjan Gudum

Annual Culture Annual Culture Annual Culture Annual Culture Annual Culture Annual Culture Sacred Grove Sacred Grove Sacred Grove Sacred Grove Sacred Grove Sacred Grove Wooded Savanna Wooded Savanna Wooded Savanna Wooded Savanna Wooded Savanna Wooded Savanna Young Secondary Forest Young Secondary Forest Young Secondary Forest Young Secondary Forest Young Secondary Forest Young Secondary Forest

619299.5087 613159.9913 617220.2954 611866.73 617265.2814 616195.1097 611938.9051 615911.8625 618618.6664 611648.131 611300.3894 615441.7503 620958.8311 618473.7045 620005.72 614308.1185 612253.2063 616543.5767 611540.741 619716.415 616063.8019 621056.3206 613829.1561 613668.6675

1306654.713 1308700.045 1308556.27 1309769.25 1312425.474 1305039.075 1312351.901 1308848.019 1310353.354 1307490.194 1305756.273 1308021.708 1305376.473 1307482.806 1305311.887 1312351.522 1308360.586 1310782.185 1310107.795 1309600.074 1305665.285 1305979.616 1311597.487 1306912.066

67.646 73.5331 61.2146 87.0459 66.3698 68.5074 87.1116 40.911 81.1985 73.3746 71.5514 92.4495 87.3181 75.6972 97.7949 94.0851 98.326 89.9869 70.9674 104.036 84.7418 66.8507 82.1682 73.8264

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

11

Table A.2 Vegetation measures used for analysis. Vertical height heterogeneity (Shannon index), Canopy cover (%), Landscape diversity (Shannon index). Measures not considered for analysis are marked with an asterisk (*): Tree density (No of Trees), Diameter at breast height (DBH) and Tree Diversity (Shannon index), based in indigenous names. Number of trees (trees/100 m2)*

DBH (DBH/ 100 m2)*

Tree Diversity/100 m2 (Shannon Index)*

52.6 ± 25.7 1.0 ± 0.1 48.0 ± 25.7 1.1 ± 0.1

7.3 ± 2.5 1 ± 1.2

30.2 ± 7.2 15.8 ± 22.7

1.1 ± 0.4 0.2 ± 0.3

30.2 ± 20.2 1.1 ± 0.1 40.1 ± 12.5 0.5 ± 0.2

0.8 ± 0.9 3.5 ± 3.0

22.5 ± 26.4 17.2 ± 9.9

0.1 ± 0.2 0.8 ± 0.9

Vegetation measure

Vertical height heterogeneity (Shannon Index)

Canopy cover (%)

Sacred Grove Young Secondary Forest Annual Culture Wooded Savanna

1.4 ± 0.2 1.2 ± 0.1

1.3 ± 1.7 1.3 ± 0.1

Landscape diversity (Shannon index)

Table A.3 Information on the studied species. The table includes the species analyzed in this study, identified by Family, English, and Scientific name. Trait categories are given for Biome association (N ¼ non-restricted, GCF ¼ Guinean-Congo forest restricted, SGS¼ Sudan-Guinean Savanna restricted), Body size (geometric mean, following Borrow & Demey, 2008), Feeding guild (c ¼ carnivorous, i ¼ insectivorous, o ¼ omnivorous, g ¼ granivorous, f ¼ frugivorous, n ¼ nectarivorous), Migratory behavior (r ¼ resident, bv ¼ breeding visitor, nb ¼ nonbreeding visitor), Habitat guild (f ¼ forest, sf ¼ bush/secondary forest, ws ¼ wooded savanna, sv ¼ savanna, mh ¼ rank herbage/moist habitats, a ¼ aerial) and IUCN population trends (unknown, stable, decreasing, increasing). Family

Common name

Scientific name

Biomeassociation

Body size (g)

Feeding guild

Migratory behavior

Habitat guild

IUCN population trend

Habitats of this study

Ardeidae

Grey Heron Purple heron Cattle Egret Hamerkop

Ardea cinerea Ardea purpurea Bubulcus ibis Scopus umbretta

N N N N

95 84 52 53

o o o c

r r r r

mh mh mh mh

unknown decreasing increasing stable

N GCF

29 25.5

c c

r r

ws f

stable decreasing

ANC, YSEC ANC ANC ANC, SAV, YSEC ANC ANC

N N

42 42.5

c c

r bv

f ws

decreasing decreasing

SAGR SAV

N N

45 60

c o

r r

ws f

increasing stable

SAV ANC

Scopidae Accipitridae

Shikra Red-thighed Sparrowhawk African Goshwak Grasshopper Buzzard

Accipiter badius Accipiter erythropus Accipiter tachiro Butastur rufipennis Red-necked Buzzard Buteo auguralis Palm-nut Vulture Gypohierax angolensis Dark Chanting Goshawk Melierax metabates Gabar Goshawk Micronisus gabar

N

43

c

r

ws

stable

SAV

N

32

c

r

ws

stable

Black Kite Hooded Vulture

N N

55 70

c c

bv r

sv ws

unknown decreasing

ANC, SAGR, SAV ANC, SAV ANC, SAV

N

64

c

r

ws

stable

ANC, SAV

N N N

34.5 29.5 12

c c g

r r r

ws ws ws

stable decreasing decreasing

ANC, SAV SAV ANC, SAV

SGS

33

g

r

sf

decreasing

ANC, YSEC

N

32.5

g

r

sv

decreasing

N

25

o

r

sv

stable

ANC, SAGR, SAV, YSEC ANC

N

21

o

r

mh

unknown

ANC, YSEC

N

32

g

r

ws

increasing

N

24

g

r

ws

stable

N

25

g

r

sv

stable

ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC

African Harrier Hawk Falconidae Fringillidae

Grey Kestrel African Hobby Yellow-fronted Canary

Phasianidae

Ahanta Francolin

Double-spurred Francolin Odontophoridae Stone Partridge Rallidae

Black Crake

Columbidae

Red-eyed Dove Laughing Dove Vinaceous Dove

Milvus migrans Necrosyrtes monachus Polyboroides typus Falco ardosiaceus Falco cuvierii Serinus mozambicus Francolinus ahantensis Francolinus bicalcaratus Ptilopachus petrosus Amaurornis flavirostra Streptopelia semitorquata Streptopelia senegalensis Streptopelia vinacea

(continued on next page)

12

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Table A.3 (continued ) Family

Psittacidae

Musophagidae

Cuculidae

Common name

Scientific name

Biomeassociation

Body size (g)

Feeding guild

Migratory behavior

Habitat guild

IUCN population trend

Habitats of this study

African Green Pigeon

Treron calvus

N

26.5

f

r

ws

decreasing

Black-billde Wood Dove Turtur abyssinicus N

20

f

r

sf

stable

Blue-spotted Wood Dove Senegal Parrot

20

g

r

ws

stable

Poicephalus SGS senegalus Rose-ringed Parakeet Psittacula krameri N Western Grey Plantain- Crinifer piscator N eater Violet Turaco Musophaga SGS violacea Green Turaco Tauraco persa GCF

23

f

r

ws

stable

ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC

40 50

f f

r r

sv ws

increasing stable

47.5

f

r

f

stable

41.5

f

r

f

stable

Black Coucal Senegal Coucal

N N

32.5 38

i i

r r

mh ws

stable stable

N

33

i

r

f

stable

SAV ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAV ANC, SAGR, SAV, YSEC ANC, SAGR

N

19

i

bv

ws

stable

ANC

N

23

i

bv

f

stable

N

18

i

r

ws

stable

N

32

i

nb

ws

stable

Turtur afer

Centropus grillii Centropus senegalensis Yellowbill Ceuthmochares aereus Didric cuckoo Chrysococcyx caprius African Emerald Cuckoo Chrysococcyx cupreus Klaas's cuckoo Chrysococcyx klaas African Cuckoo Cuculus gularis Red-chested Cuckoo

Cuculus solitarius N

29.5

i

bv

f

stable

Levaillant's Cuckoo

Oxylophus levaillantii Bubo africanus cinerascens Apus affinis Apus apus Cypsiurus parvus Telacanthura ussheri Ceyx pictus

N

39

i

bv

ws

stable

ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAGR, SAV, YSEC ANC, SAV

N

45.5

c

r

ws

stable

SAV

N N N N

12.75 17.75 16 13.5

i i i i

r nb r r

a a ws ws

increasing stable increasing stable

ANC, SAV ANC, SAV ANC, SAV ANC

N

12

i

r

sf

stable

N N

17 22

c c

r bv

ws ws

stable stable

ANC, SAGR, SAV, YSEC SAV ANC, SAV

N

25

c

r

f

decreasing

N

22

c

r

ws

stable

ANC, SAGR, SAV, YSEC ANC, YSEC

N N

24 29

i i

nb r

ws ws

stable stable

SAV SAV

SGS

29

i

r

ws

decreasing

N

29.5

i

r

ws

stable

ANC, SAV, YSEC ANC, SAV

N

37.5

i

r

ws

decreasing

N

23

i

r

ws

decreasing

N

48

o

r

ws

stable

N

51.5

o

r

f

unknown

N

20

i

r

f

stable

GCF

15

i

r

f

stable

Strigidae

Greyish Eagle Owl

Apodidae

Little Swift Common Swift African Palm Swift Mottled Spinetail

Alcedinidae

African Pygmy Kingfisher Striped Kingfisher Halcyon chelicuti Grey-headed Kingfisher Halcyon leucocephala Blue-breasted Kingfisher Halcyon malimbica Woodland kingfisher Halcyon senegalensis European Bee-eater Merops apiaster Abyssinian Roller Coracias abyssinicus Blue-bellied Roller Coracias cyanogaster Broad-billed Roller Eurystomus glaucurus Green Wood Hoopoe Phoeniculus purpureus Black Scimitarbill Rhinopomastus aterrimus African Grey Hornbill Lophoceros nasutus African Pied Hornbill Lophoceros semifasciatus Golden-tailed Campethera Woodpecker abingoni Buff-spotted Campethera Woodpecker nivosa

Meropidae Coraciidae

Phoeniculidae

Bucerotidae

Picidae

N

ANC, SAV, YSEC ANC, SAV ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAGR, SAV, YSEC SAGR

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

13

Table A.3 (continued ) Family

Common name

Scientific name

Biomeassociation

Body size (g)

Feeding guild

Migratory behavior

Habitat guild

IUCN population trend

Habitats of this study

Fine-spotted Woodpecker Cardinal Woodpecker

Campethera punctuligera Dendropicos fuscescens Dendropicos goertae Picoides obsoletus

N

21

i

r

ws

stable

ANC, SAV

N

14.5

i

r

ws

stable

ANC, SAV

N

20

i

r

ws

stable

N

13.5

i

r

ws

stable

ANC, SAV, YSEC ANC, SAV

N

13

o

r

sf

decreasing

SAGR, YSEC

SGS N

14 13.5

g g

r r

sv sv

stable decreasing

SAV SAV

N

17

i

bv

ws

increasing

ANC, SAV

SGS

12

i

r

ws

increasing

ANC, SAV

N GCF

15 17

i i

r r

ws sf

increasing stable

ANC ANC, SAV

N N

12.5 20

i i

nb r

a f

decreasing decreasing

GCF

16

f

r

f

stable

ANC ANC, SAV, YSEC SAV

N

16.5

o

r

sf

stable

SAGR

GCF

21.25

i

r

f

stable

N

22.5

f

r

sf

stable

GCF

21

o

r

sf

stable

ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC SAGR, YSEC

GCF N

22.5 19

i o

r r

f ws

stable increasing

GCF

22

i

r

f

stable

N

23

o

r

f

unknown

SGS N N

10 12.5 19.5

i i i

r r r

ws ws sv

stable decreasing stable

N

11.5

i

r

sf

increasing

N

11.5

i

r

ws

stable

N N N

12.75 13 13.5

i i i

r r r

sf ws mh

stable stable stable

SGS N N

10 13.25 12

i i i

r r r

ws ws ws

stable stable stable

SGS

20

i

r

sf

stable

N

11.5

i

r

ws

stable

16 26

i i

r r

ws sf

stable stable

21.25

i

r

sf

stable

20

i

r

ws

stable

Grey Woodpecker

Eurylaimidae

Brown-backed Woodpecker African Broadbill

Smithornis capensis Alaudidae Sun Lark Galerida modesta Flappet Lark Mirafra rufocinnamomea Hirundinidae Lesser-striped Swallow Hirundo abyssinica Pied-winged Swallow Hirundo leucosoma Red-chested Swallow Hirundo lucida Fanti Saw-Wing Psalidoprocne obscura Sand Martin Riparia riparia Campephagidae Red-shouldered Cuckoo- Campephaga Shrike phoenicea Pycnonotidae Ansorge's Greenbul Andropadus ansorgei Little Greenbul Andropadus virens Grey-headed Bristlebill Bleda canicapillus Yellow-throated Leaflove Simple Leaflove

Turdidae

African Thrush

Chlorocichla flavicollis Chlorocichla simplex Nicator chloris Pycnonotus barbatus Pyrrhurus scandens Turdus pelios

Sylviidae

Senegal Eremomela Yellow-bellied Hyliota African Moustached Warbler Yellow-breasted Apalis

Eremomela pusilla Hyliota flavigaster Melocichla mentalis Apalis flavida

Grey-backed Camaroptera Singing Cisticola Red-faced Cisticola Winding Cisticola

Camaroptera brachyura Cisticola cantans Cisticola erythrops Cisticola galactotes Cisticola rufus Citicola lateralis Heliolais erythropterus Hypergerus atriceps Prinia subflava

Western Nicator Common Bulbul Leaflove

Cisticolidae

Rufous Cisticola Whistling Cisticola Red-winged Warbler Oriole Warbler Tawny-flanked Prinia Muscicapidae

Pale Flycatcher White-crowned Robin Chat Snowy-Crowned Robin Chat

Bradornis pallidus N Cossypha SGS albicapilla Cossypha N niveicapilla N

SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC SAV ANC, SAV ANC ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAV ANC SAV ANC, YSEC ANC ANC, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAV ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC

(continued on next page)

14

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Table A.3 (continued ) Family

Platysteiridae

Timaliidae

Common name

Scientific name

Northern Black Flycatcher Red-bellied Paradise Flycatcher African ParadiseFlycatcher Senegal Batis Common Wattle-eye

Melaenornis edolioides Terpsiphone rufiventer Terpsiphone viridis Batis senegalensis Platysteira cyanea

Brown Illadopsis

Illadopsis GCF fulvescens Phyllanthus GCF atripennis Turdoides plebejus N

decreasing

N

18

i

r

sf

stable

N N

10.5 13

i i

r r

ws sf

decreasing stable

16

i

r

f

stable

24

o

r

sf

unknown

24

o

r

sf

stable

SGS

25

o

r

mh

decreasing

N

14

o

r

sv

decreasing

N

10.5

i

r

ws

stable

N

14.5

fv

r

f

stable

N

13.5

fv

r

ws

stable

SGS

14

fv

r

ws

stable

N

12.5

fv

r

ws

stable

N

10

fv

r

ws

stable

N N

10 14

fv fv

r r

ws f

stable stable

N

13.5

fv

r

ws

stable

N

10

fv

r

sf

stable

N

9.5

fv

bv

ws

stable

SGS N

31.5 15

i f

r r

ws ws

unknown unknown

N

10

f

r

f

stable

N

11.5

f

r

ws

stable

N

18.75

i

r

ws

increasing

GCF

17

o

r

f

decreasing

N

14.5

o

r

ws

stable

Wahlbergs's Honeybird Prodotiscus regulus Northern Puffback Dryoscopus gambensis Yellow-crowned Laniarius Gonolek barbarus Turatis Boubou Laniarius turatii

N

12.5

i

r

sf

increasing

N

18.5

i

r

ws

stable

N

23

i

r

sf

stable

GCF

23

i

r

sf

stable

Grey-headed Bushshrike Malaconotus blanchoti Brubru Nilaus afer White helmet-Shrike Prionops plumatus Black-crowned Tchagra Tchagra senegalus

N

25

i

r

ws

increasing

N N

14 21

i i

r r

ws ws

stable stable

N

21

i

r

ws

stable

Turdoides reinwardtii White-shouldered Black Parus guineensis Tit Yellow White-eye Zosterops senegalensis Scarlet-chested Sunbird Chalcomitra senegalensis Western violet-backed Anthreptes Sunbird longuemarei Splendid Sunbird Cinnyris coccinigastrus Copper Sunbird Cinnyris cupreus

Green-headed Sunbird Collared Sunbird Pygmy Sunbird Yellow-billed Shrike Vieillot's Barbet Yellow-rumped Tinkerbird Yellow-fronted Tinkerbird Greater Honeyguide Spotted Honeyguide Lesser Honeyguide

Malaconotidae

IUCN population trend

sf

Variable Sunbird Olive Sunbird

Indicatoridae

Habitat guild

r

Beautiful Sunbird

Laniidae

Migratory behavior

i

Blackcap Babbler

Nectariniidae

Feeding guild

18

Brown Babbler

Zosteropidae

Body size (g)

GCF

Capuchin Babbler

Paridae

Biomeassociation

Cinnyris pulchellus Cinnyris venustus Cyanomitra olivacea Cyanomitra verticalis Hedydipna collaris Hedydipna platura Corvinella corvina Lybius vieilloti Pogoniulus bilineatus Pogoniulus chrysoconus Indicator indicator Indicator maculatus Indicator minor

Habitats of this study

ANC, SAV, YSEC ANC, SAGR, SAV, YSEC SAGR, SAV, YSEC ANC, SAV ANC, SAGR, SAV, YSEC SAGR, SAV, YSEC ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAV ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAV ANC, SAGR, SAV, YSEC ANC, SAV, YSEC ANC, SAV, YSEC ANC, SAV SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV ANC, SAV ANC, SAGR, SAV ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAV, YSEC ANC, SAGR, YSEC ANC, SAV, YSEC ANC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAV ANC, SAV, YSEC

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

15

Table A.3 (continued ) Family

Oriolidae

Common name

Sulphur-breasted Bush- Telophorus Shrike sulfureopectus African Golden Oriole Oriolus auratus Black-winged Oriole

Dicruridae

Corvidae Sturnidae

Passeridae

Ploceidae

Eurasian Golden Oriole Fork-tailed Drongo

Biomeassociation

Body size (g)

Feeding guild

Migratory behavior

Habitat guild

IUCN population trend

Habitats of this study

ANC, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC ANC, SAGR, SAV, YSEC SAV ANC, SAGR, SAV, YSEC SAGR SAV ANC

N

18

i

r

ws

stable

N

24

o

r

ws

stable

20

o

r

f

decreasing

24 23.75

o i

nb r

f ws

stable stable

N SGS N

19 35 22

i o o

r r r

sf ws ws

stable stable decreasing

N

17

o

bv

ws

decreasing

N

21.5

o

r

ws

stable

ANC, SAGR, SAV, YSEC ANC, SAV

N

19.5

o

r

ws

stable

SAV

SGS

24

o

r

sv

stable

ANC

N

14

g

r

ws

stable

ANC, SAV

N SGS

13 17

g g

r r

ws ws

stable stable

SAV SAV

N

11.5

g

r

mh

stable

SAV

N

13.5

i

r

mh

stable

ANC

Oriolus GCF nigripennis Oriolus oriolus N Dicrurus adsimilis N

Square-tailed Drongo Dicrurus ludwigii Piapiac Ptilostomus afer Yellow-billed Oxpecker Buphagus africanus Violet-backed Starling Cinnyricinclus leucogaster Bronze-tailed Glossy Lamprotornis Starling chalcurus Lesser Blue-eared Lamprotornis Starling chloropterus Purple Glossy Starling Lamprotornis purpureus Northern Grey-headed Passer griseus Sparrow Bush Petronia Petronia dentata Chestnut-crowned Plocepasser Sparrow Weaver superciliosus Northern Red Bishop Euplectes franciscanus Back-winged Bishop Euplectes hordeaceus Yellow-mantled Euplectes Widowbird macroura Village Weaver Ploceus cucullatus

N

16.75

g

r

mh

stable

ANC, SAV

N

16.25

g

r

ws

stable

Little Weaver

Ploceus luteolus

N

11.5

g

r

ws

stable

Black-headed Weaver

Ploceus N melanocephalus Ploceus nigricollis N

14.5

o

r

mh

stable

17

o

r

ws

stable

Ploceus vitellinus

N

14

g

r

sv

stable

Quelea quelea Estrilda caerulescens Estrilda melpoda

N SGS

12 10

g g

r r

ws ws

stable stable

N

10

g

r

sv

stable

ANC, SAGR, SAV, YSEC ANC, SAV, YSEC ANC, SAV, YSEC ANC, SAGR, YSEC ANC, SAGR, SAV, YSEC ANC, SAV ANC, SAGR, YSEC ANC, SAV

Euschistospiza dybowskii Lagonosticta larvata Lagonosticta rara Lagonosticta rubricata Lagonosticta rufopicta Lagonosticta senegala Lonchura cucullata Pyrenestes sanguineus Pytilia phoenicoptera

SGS

12

g

r

ws

stable

ANC, YSEC

SGS

11.5

g

r

ws

stable

ANC

SGS N

10 10

g g

r r

sv ws

stable stable

ANC, SAV ANC, YSEC

SGS

11

g

r

sv

stable

ANC

N

10

g

r

ws

stable

ANC, YSEC

N

9

g

r

sf

stable

ANC, YSEC

GCF

13.5

g

r

f

stable

SAGR

SGS

12.75

g

r

ws

stable

ANC, SAV

Back-necked Weaver

Estrildidae

Scientific name

Vitelline-masked Weaver Red-billed Quelea Lavender Waxbill Orange-cheeked Waxbill Dybowski's Twinspot Black-faced Firefinch Black-bellied Firefinch Blue-billed Firefinch Bar-breasted Firefinch Red-billed Firefinch Bronze Mannikin Crimson Seedcracker Red-winged Pytilia

(continued on next page)

16

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Table A.3 (continued ) Family

Common name

Scientific name

Biomeassociation

Body size (g)

Feeding guild

Migratory behavior

Habitat guild

IUCN population trend

Habitats of this study

Uraeginthus bengalus Vidua interjecta

N

13

g

r

sv

stable

ANC, SAV

Viduidae

Red-cheeked CordonBleu Exclamatory ParadiseWhydah Pin-tailed Whydah

N

12.5

g

r

ws

stable

SAV, YSEC

Vidua macroura

N

12.5

g

r

ws

stable

ANC, SAV

Fig. A.1. Species accumulation curves per habitat type (SAGR¼ Sacred Groves, YSEC¼ Young Secondary Forests, SAV¼Wooded Savannas, and ANC ¼ Annual Cultures).

References Aerts, R., Van Overtveld, K., Haile, M., Hermy, M., Deckers, J., Muys, B., 2006. Species composition and diversity of small Afromontane forest fragments in northern Ethiopia. Plant Ecol. 187 (1), 127e142. Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecol. 26 (1), 32e46. Arcilla, N., Holbech, L.H., O'Donnell, S., 2015. Severe declines of understory birds follow illegal logging in Upper Guinea forests of Ghana, West Africa. Biol. Conserv. 188, 41e49. Barre, R.Y., Grant, M., Draper, D., 2009. The role of taboos in conservation of sacred groves in Ghana's Tallensi-Nabdam district. Soc. Cult. Geogr. 10 (1), 25e39. Bender, D.J., Contreras, T.A., Fahrig, L., 1998. Habitat loss and population decline: a meta-analysis of the patch size effect. Ecology 79 (2), 517e533. Bhagwat, S.A., Rutte, C., 2006. Sacred groves: potential for biodiversity management. Front. Ecol. Environ. 4 (10), 519e524. Bhagwat, S.A., Kushalappa, C.G., Williams, P.H., Brown, N.D., 2005. A landscape approach to biodiversity conservation of sacred groves in the Western Ghats of India. Conserv. Biol. 19 (6), 1853e1862. Bibby, C.J., Burgess, N.D., Hill, D.A., Mustoe, S., 2000. Bird census techniques. Elsevier, London, UK. Binczik, A., Roig-Boixeda, P., Heymann, E.W., Waltert, M., 2017. Conservation of chimpanzees Pan troglodytes verus and other primates depends on forest patches in a West African savannah landscape. Oryx 1e8. Borrow, N., Demey, R., 2008. Birds of Western Africa. Princeton University Press, London, UK. Borrow, N., Demey, R., 2011. Birds of Senegal and the Gambia. Christopher Helm Field Guides, London, UK. Brandt, J.S., Butsic, V., Schwab, B., Kuemmerle, T., Radeloff, V.C., 2015. The relative effectiveness of protected areas, a logging ban, and sacred areas for oldgrowth forest protection in southwest China. Biol. Conserv. 181, 1e8. Brandt, J.S., Wood, E.M., Pidgeon, A.M., Han, L.X., Fang, Z., Radeloff, V.C., 2013. Sacred forests are keystone structures for forest bird conservation in southwest China's Himalayan Mountains. Biol. Conserv. 166, 34e42. Brink, A.B., Eva, H.D., 2009. Monitoring 25 years of land cover change dynamics in Africa: a sample based remote sensing approach. Appl. Geogr. 29 (4), 501e512. Brooks, T.M., Pimm, S.L., Oyugi, J.O., 1999. Time lag between deforestation and bird extinction in tropical forest fragments. Conserv. Biol. 13 (5), 1140e1150. Brooks, T.M., Mittermeier, R.A., Mittermeier, C.G., Da Fonseca, G.A., Rylands, A.B., Konstant, W.R., Flick, P., Pilgrim, J., Oldfield, S., Magin, G., Hilton-Taylor, C., 2002. Habitat loss and extinction in the hotspots of biodiversity. Conserv. Biol. 16 (4), 909e923. Campbell, M.O.N., 2005. Sacred Groves for forest conservation in Ghana's coastal savannas: assessing ecological and social dimensions. Singapore J. Trop. Geogr. 26 (2), 151e169. Ceperley, Natalie, Montagnini, Florencia, Armand, Natta, 2010. Significance of sacred sites for riparian forest conservation in Central Benin. Bois Forets Tropiques 303, 5e23. ARTICLE. Chao, A., 1987. Estimating the Population Size for Capture-Recapture Data with Unequal Catchability. Biometrics, pp. 783e791. Chimbo Foundation, 2019. COMBAC Boe accessed May 2019. http://chimbo.org/?page_id¼975page_id¼62&lang¼en. Colwell, R.K., Coddington, J.A., 1994. Estimating terrestrial biodiversity through extrapolation. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 345 (1311), 101e118. Decher, J., 1997. Conservation, small mammals, and the future of sacred groves in West Africa. Biodivers. Conserv. 6 (7), 1007. Dray, S., Dufour, A.B., 2007. The ade4 package: implementing the duality diagram for ecologists. J. Stat. Softw. 22 (4), 1e20. Dray, S., Legendre, P., 2008. Testing the species traitseenvironment relationships: the fourth-corner problem revisited. Ecology 89 (12), 3400e3412. Fishpool, L.D., Evans, M.I., 2001. Important Bird Areas in Africa and Associated Islands: Priority Sites for Conservation. Pisces Publications and BirdLife International (Birdlife Conservation Series no. 11), Newbury and Cambridge, UK. Freedman, A.H., Buermann, W., Mitchard, E.T., DeFries, R.S., Smith, T.B., 2010. Human impacts flatten rainforest-savanna gradient and reduce adaptive diversity in a rainforest bird. PLoS One 5 (9), e13088. Fry, C.H., Fry, K., 2010. Kingfishers, Bee-Eaters and Rollers. Cristopher Helm Publishers Ltd, London, UK.

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

17

Gao, H., Ouyang, Z., Chen, S., Van Koppen, C.S.A., 2013. Role of culturally protected forests in biodiversity conservation in Southeast China. Biodivers. Conserv. 22 (2), 531e544. € rsch, B., Porembski, S., 2006. Dynamics of forestesavanna mosaics in north-eastern Ivory Coast from 1954 to 2002. J. Biogeogr. 33 (4), Goetze, D., Ho 653e664. Gray, M.A., Baldauf, S.L., Mayhew, P.J., Hill, J.K., 2007. The response of avian feeding guilds to tropical forest disturbance. Conserv. Biol. 21 (1), 133e141. Gregory, S.V., Swanson, F.J., McKee, W.A., Cummins, K.W., 1991. An ecosystem perspective of riparian zones. Bioscience 41 (8), 540e551. Helms, M.W., 2002. Sacred Landscape and the Early Medieval European Cloister. Unity, Paradise, and the Cosmic Mountain. Anthropos, pp. 435e453. IUCN, 2015. The IUCN List of Threatened Species. Version2015-4. http://www.iucnredlist.org. December 2015. Jalloh, A., Roy-Macauley, H., Sereme, P., 2012. Major agro-ecosystems of West and Central Africa: brief description, species richness, management, environmental limitations and concerns. Agric. Ecosyst. Environ. 157, 5e16. Kinley, T.A., Newhouse, N.J., 1997. Relationship of riparian reserve zone width to bird density and diversity in southeastern British Columbia. Northwest Sci. 71, 75e86. Lebbie, A.R., Freudenberger, M.S., 1996. Sacred groves in Africa: forest patches in transition. Forest patches in tropical landscapes 300e324. Lymer, K., 2004. Rags and rock art: the landscapes of holy site pilgrimage in the Republic of Kazakhstan. World Archaeol. 36 (1), 158e172. MacArthur, R.H., MacArthur, J.W., 1961. On bird species diversity. Ecology 42 (3), 594e598. Maconachie, R., Srinivasan, R., Menzies, N., 2015. Responding to the Challenge of Fragility and Security in West Africa. March 2019. https://openknowledge. worldbank.org/handle/10986/22511. Mallon, D.P., Hoffmann, M., Grainger, M.J., Hibert, F., Van Vliet, N., McGowan, P.J.K., 2015. An IUCN situation analysis of terrestrial and freshwater fauna in West and Central Africa. Occas. Pap. IUCN Species Surviv. Comm. 54, 40. Mandal, J., Shankar Raman, T.R., 2016. Shifting agriculture supports more tropical forest birds than oil palm or teak plantations in Mizoram, northeast India. Condor 118 (2), 345e359. € hning-Gaese, K., 2012. Meta-analysis of the effects of human Markl, J.S., Schleuning, M., Forget, P.M., Jordano, P., Lambert, J.E., Traveset, A., Wright, S.J., Bo disturbance on seed dispersal by animals. Conserv. Biol. 26 (6), 1072e1081. Møller, A.P., Rubolini, D., Lehikoinen, E., 2008. Populations of migratory bird species that did not show a phenological response to climate change are declining. Proc. Natl. Acad. Sci. Unit. States Am. 105 (42), 16195e16200. Monteiro, F., Catarino, L., Batista, D., Indjai, B., Duarte, M., Romeiras, M., 2017. Cashew as a high agricultural commodity in West Africa: insights towards sustainable production in Guinea-Bissau. Sustainability 9 (9), 1666. Morelli, F., Pruscini, F., Santolini, R., Perna, P., Benedetti, Y., Sisti, D., 2013. Landscape heterogeneity metrics as indicators of bird diversity: determining the optimal spatial scales in different landscapes. Ecol. Indicat. 34, 372e379. Naidoo, R., 2004, January. Species richness and community composition of songbirds in a tropical forest-agricultural landscape. In: Animal Conservation Forum, vol. 7. Cambridge University Press, pp. 93e105. No. 1. Nichols, J.D., Conroy, M.J., 1996. Estimation of species richness. In: Wilson, D.E., Cole, F.R., Nichols, J.D., Rudran, R., Foster, M. (Eds.), Measuring and Monitoring Biological Diversity. Standard Methods for Mammals. Smithsonian Institution Press, Washington, United States of America, pp. 226e234. Ogada, D.L., Keesing, F., Virani, M.Z., 2012. Dropping dead: causes and consequences of vulture population declines worldwide. Ann. N. Y. Acad. Sci. 1249 (1), 57e71. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, L.G., Solymos, P., Henry, M., Stevens, H., Wagner, H., 2015. Vegan: Community Ecology Package. R Package Version 2.3-2. Ormsby, A.A., Bhagwat, S.A., 2010. Sacred forests of India: a strong tradition of community-based natural resource management. Environ. Conserv. 37 (3), 320e326. Oteng-Yeboah, A.A., 1996. Biodiversity in three traditional groves in the Guinea savanna, Ghana. In: Van der Maesen, L.J.G., van der Burgt, X.M., van Medenbach de Rooy, J.M. (Eds.), The Biodiversity of African Plants. Springer, Dordrecht, Netherlands, pp. 188e197. Palmer, M.W., 1990. The estimation of species richness by extrapolation. Ecology 71 (3), 1195e1198. R Core Team, 2015. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.Rproject.org/. Raman, T.S., Rawat, G.S., Johnsingh, A.J.T., 1998. Recovery of tropical rainforest avifauna in relation to vegetation succession following shifting cultivation in Mizoram, north-east India. J. Appl. Ecol. 35 (2), 214e231. Ray, R., Chandran, M.D.S., Ramachandra, T.V., 2015. Hydrological importance of sacred forest fragments in Central Western Ghats of India. Trop. Ecol. 56 (1), 87e99. Rosenzweig, M.L., 1995. Species Diversity in Space and Time. Cambridge University Press, UK. Sanderson, F.J., Donald, P.F., Pain, D.J., Burfield, I.J., Van Bommel, F.P., 2006. Long-term population declines in Afro-Palearctic migrant birds. Biol. Conserv. 131 (1), 93e105. S¸ekercioḡlu, Ç.H., Ehrlich, P.R., Daily, G.C., Aygen, D., Goehring, D., Sandí, R.F., 2002. Disappearance of insectivorous birds from tropical forest fragments. Proc. Natl. Acad. Sci. Unit. States Am. 99 (1), 263e267. Selmi, S., Boulinier, T., Barbault, R., 2002. Richness and composition of oasis bird communities: spatial issues and speciesearea relationships. Auk 119 (2), 533e539. Shirley, S.M., Smith, J.N., 2005. Bird community structure across riparian buffer strips of varying width in a coastal temperate forest. Biol. Conserv. 125 (4), 475e489. Slabbekoorn, H., Smith, T.B., 2002. Habitat-dependent song divergence in the little greenbul: an analysis of environmental selection pressures on acoustic signals. Evolution 56 (9), 1849e1858. Smith, T.B., Wayne, R.K., Girman, D.J., Bruford, M.W., 1997. A role for ecotones in generating rainforest biodiversity. Science 276 (5320), 1855e1857. € € m, B., Kiema, S., Reid, R.S., 2003. Intensified agricultural land-use and bird conservation in Burkina Faso. Agric. Ecosyst. Environ. 99 (1e3), 113e124. Soderstro Sreevidya, E.A., Pattabh, S., Ray, Rajasri, 2016. Conservation in heterogeneous landscape - sacred groves matter. IOSR J. Environ. Sci. Toxicol. Food Technol. 10, 2319e2399. Sutherland, W.J., 2006. Ecological Census Techniques: a Handbook. Cambridge university press, UK. Temudo, M.P., 2012. The white men bought the forests": conservation and contestation in Guinea-Bissau, western Africa. Conserv. Soc. 10 (4), 354. Temudo, M.P., Abrantes, M., 2014. The cashew frontier in Guinea-Bissau, West Africa: changing landscapes and livelihoods. Hum. Ecol. 42 (2), 217e230. Temudo, M.P., Santos, P., 2017. Shifting environments in Eastern Guinea-Bissau, West Africa: the length of fallows in question. NJAS - Wageningen J. Life Sci. 80, 57e64. € rger, K., Wichmann, M.C., Schwager, M., Jeltsch, F., 2004. Animal species diversity driven by habitat heterogeneity/ Tews, J., Brose, U., Grimm, V., Tielbo diversity: the importance of keystone structures. J. Biogeogr. 31 (1), 79e92. Thiollay, J.M., 2006a. The decline of raptors in West Africa: long-term assessment and the role of protected areas. Ibis 148 (2), 240e254. Thiollay, J.M., 2006b. Large bird declines with increasing human pressure in savanna woodlands (Burkina Faso). Biodivers. Conserv. 15 (7), 2085. Trail, P.W., 2007. African hornbills: keystone species threatened by habitat loss, hunting and international trade. Ostrich - J. Afr. Ornithol. 78 (3), 609e613. Tscharntke, T., Sekercioglu, C.H., Dietsch, T.V., Sodhi, N.S., Hoehn, P., Tylianakis, J.M., 2008. Landscape constraints on functional diversity of birds and insects in tropical agroecosystems. Ecology 89 (4), 944e951.  , J., Gregory, R.D., 2014. The decline of Afro-Palaearctic migrants and an assessment of Vickery, J.A., Ewing, S.R., Smith, K.W., Pain, D.J., Bairlein, F., Skorpilov a potential causes. Ibis 156 (1), 1e22.

18

K. Kühnert et al. / Global Ecology and Conservation 18 (2019) e00656

Vinay, S., Vishnu, D.M., Srikanth, N., Asulabha, K.S., Sincy, V., Rao, G.R., Shashishankar, A., Chandran, M.D.S., Ramachandra, T.V., 2017. Hydrological Regime in Sacred Groves and Non-sacred Groves of Central Western Ghats. March 2019. http://wgbis.ces.iisc.ernet.in/energy/lake2016/EWRG/T1_12_Vinay% 20LAKE%202016.pdf. Waltert, M., Bobo, K.S., Sainge, N.M., Fermon, H., Mühlenberg, M., 2005. From forest to farmland: habitat effects on Afrotropical forest bird diversity. Ecol. Appl. 15 (4), 1351e1366. White, F., 1983. Vegetation of Africa - a Descriptive Memoir to Accompany the Unesco/AETFAT/UNSO Vegetation Map of Africa. Natural Resources Research Report XX. U.N. Educational, Scientific and Cultural Organization, Paris, France. Whytock, R.C., Morgan, B.J., 2010. The commercial trade in bushmeat potentially threatens raptor populations in the Ebo forest, Cameroon. Gabar 21, 1e7. Whytock, R.C., Buij, R., Virani, M.Z., Morgan, B.J., 2016. Do large birds experience previously undetected levels of hunting pressure in the forests of Central and West Africa? Oryx 50 (1), 76e83. Wiens, J., Johnston, R., 2012. Adaptive correlates of granivory in birds. In: Pinowski, S., Kendeigh, C. (Eds.), Granivorous Birds in Ecosystems: Their Evolution, Populations, Energetics, Adaptations, Impact and Control. Cambridge University Press, Cambridge, United Kingdom, pp. 301e340. Willson, M.F., 1974. Avian community organization and habitat structure. Ecology 55 (5), 1017e1029.  province, Guinea Bissau. Agric. Ecosyst. Environ. 27 (1e4), 609e620. Wit, P., Reintjes, H.C., 1989. An agro-ecological survey of the Boe